The separation nut full, half full and empty non-destructive method using fuzzy logic
Subject Areas :
Keywords: fuzzy logic, Calibration, Walnut, sound processing, nondestructive testing,
Abstract :
Walnut is one of the most important horticultural crops around the world. It also plays an important role in agriculture of Iran. One reason for the small amount of export Persian walnut, uniform product is to use different varieties of the reasons for One of the best solutions to eliminate the lack of uniformity in the export of goods Walnut, grading walnut or walnut shell (shell apart) using a nondestructive method is. The overall objective of this study was to isolate nut full, half full and empty, without breaking them, based on analysis methods and fuzzy logic was sound. Walnut separation using acoustic sound processing approach was used. In this context the isolation and detection of the desired quality parameters were studied. Some of these parameters can be geometric features, size and shape of a walnut cited. For this purpose, 300 Walnut for use in experiments randomly selected from a garden walnuts and walnut. In this stage each of the three for the back, sides and belly with a height of 30 cm from the surface of the voice box, was released, The more acoustic sounds by tapping sound chamber made with wood surface and a slope of 45 degrees was recorded by the microphone and get in it. The data stored in computers and coding of audio signals in the time domain was done in MATLAB software programming environment, Then were analyzed by fuzzy logic. The results of this study showed that, in the classification of walnut, decision trees because of their simplicity and structure of fuzzy rules and threshold values of membership functions, it is easy to Fuzzy Inference System with high precision. Fuzzy final version submitted for classification of walnut to 2 classes 3 classes of accuracy 087/0 per cent of isolation and separation was achieved with an accuracy of 80 0/0 percent.